Lung carcinoma continues to account for more cancer deaths than any other tumor due to lack of effective screening and early detection tests and to a propensity for early metastatic spread. During the coming funding period, the Colorado EDRN Biomarker Development Laboratory will continue the phased development of molecular biomarkers to: 1) detect prevalent lung carcinoma at an early stage; 2) identify premalignant lesions in high risk smokers that may be targeted for chemoprevention; and 3) develop biomarkers that predict incident lung carcinoma in high risk smokers initially without tumor. The Laboratory will build on past success using state-of-the art oligonucleotide microarray technology to discover molecular markers that are highly overexpressed in tumors but silent in normal tissue. So far the most promising of the markers identified this method are CTAG genes which represent a unique new family of biomarker molecules that are expressed exclusively by tumor cells and germ cells. To identify small numbers of heterogeneous tumor cells in clinical samples, panels of biomarkers will be assessed by multiplex RT-PCR, real time quantitative RT-PCR and protein immunoassays in a broad range of cell lines, tumors, normal tissues, peripheral blood, sputum and other surrogate markers. An underlying premise of the program is that unbalance gene expression derives from chromosomal instability. Chromosomal instability is reflected by aneusomy in malignant and premalignant epithelial cells. We plan to validate chromosomal instability/aneusomy by applying multiprobe FISH to the unique retrospective repository specimens from the Colorado High Risk Cohort and to compare predictive power of FISH with cytology and methylation of suppressor gone promoters in collaboration with Dr. Steve Belinsky. We expect that one or more of these biomarkers will be sufficiently promising that it may be applied to specimens currently being collected through the spiral CT project of ACRIN currently being housed in the Colorado repository.